The frequency of therapy-related acute myeloid leukemia (t-AML) continues to increase as the cure rates for a wide-variety of cancers improves. Unfortunately, t-AML is a highly aggressive disease and is rarely curable, with median overall survival rates using current therapies of less than 10%. Detailed cytogenetic and molecular studies of t-AML, in part supported by this program project grant, have provided important insights into the underlying molecular pathogenesis of this disease including the loss of the long arm or total loss of chromosomes 5 [-5/del(5q)] and/or 7 [-7/del(7q)] in the majority of patients, and the frequent mutation of AML1/RUNX1, FLT3, MLL, NRAS, cKIT, and p53. However, the full complement of cooperating lesions responsible for the development of t-AML remains to be defined. It is our belief that defining at a molecular level the total complement of alterations that contribute to the development of t- AML will not only enhance our ability to accurately diagnosis this disease, but should also help to define the molecular "Achilles heels" against which targeted therapy can be developed. The long-term goal of this project is to identify the complement of genetic alterations that occur in t-AML. This objective will be pursued through two specific aims: (1) To identify somatically acquired genetic copy number alterations t- AML and to elucidate their mechanistic contribution to cellular transformation;and (2) To identify somatically acquired sequence mutations in a selected subset of cancer related genes. The latter will include known cancer genes and genes identified as either having copy number alterations or altered expression profiles. These specific aims will be pursued using a combination of high resolution genomewide copy number analysis and targeted resequencing on a cohort of over 200 t-AML samples obtained through the Patient Access, Data Management, Statistical Analysis, and Tissue Culture Core (Core A). Copy number analysis will be performed using Affymetrix Genome-Wide human SNP Array 6.0, which provides a resolution of <5 kb. Importantly, matched germ line samples are either available or will be generated by expansion of BM stromal fibroblasts or reactive T-cells for every patient, allowing us to definitively determine if an identified copy number change is somatically acquired. Gene mutations that are identified as the targets of recurrent copy number alterations will be confirmed using FISH or genomic quantitative PCR, and will be sequenced to identify the presence of any point mutations. The identified genes will then be directly assessed for their contribution to cellular transformation by evaluating their effect on normal hematopoietic stem cell growth and differentiation, and on their ability to cooperate with known oncogenic lesions to induce leukemia.